This report describes two wind turbine models developed within the second work package (WP2) of IEA Wind Task 37 on Wind Energy Systems Engineering: Integrated RD&D. The wind turbine models can be used as references for future research projects on wind energy, representing a modern land-based wind turbine and a latest generation offshore wind turbine. The land-based design is a class IIIA geared configuration with a rated electrical power of 3.4-MW, a rotor diameter of 130 m, and a hub height of 110 m. The offshore design is a class IA configuration with a rated electrical power of 10.0 MW, a rotor diameter of 198 m, and a hub height of 119 m. The offshore turbine employs a direct-drive generator.

Abstract.Coupling between bending and twist has a significant influence on the aeroelastic response of wind turbine blades. The coupling can arise from the blade geometry (e.g. sweep, prebending, or deflection under load) or from the anisotropic properties of the blade material. Bend-twist coupling can be utilized to reduce the fatigue loads of wind turbine blades. In this study the effects of material-based coupling on the aeroelastic modal properties and stability limits of the DTU 10 MW Reference Wind Turbine are investigated. The modal properties are determined by means of eigenvalue analysis around a steady-state equilibrium using the aero-servo-elastic tool HAWCStab2 which has been extended by a beam element that allows for fully coupled cross-sectional properties. Bend-twist coupling is introduced in the cross-sectional stiffness matrix by means of coupling coefficients that introduce twist for flapwise (flap-twist coupling) or edgewise (edge-twist coupling) bending. Edge-twist coupling can increase or decrease the damping of the edgewise mode relative to the reference blade, depending on the operational condition of the turbine. Edge-twist to feather coupling for edgewise deflection towards the leading edge reduces the inflow speed at which the blade becomes unstable. Flap-twist to feather coupling for flapwise deflections towards the suction side increase the frequency and reduce damping of the flapwise mode. Flap-twist to stall reduces frequency and increases damping. The reduction of blade root flapwise and tower bottom fore-aft moments due to variations in mean wind speed of a flap-twist to feather blade are confirmed by frequency response functions.

No abstract

Efficient ultimate load estimation for offshore wind turbines using interpolating surrogate models L M M van den Bos, B Sanderse, L Blonk et al. Wind turbine site-specific load estimation using artificial neural networks calibrated by means of high-fidelity load simulations Abstract. Previous studies have suggested the use of reduced-order models calibrated by means of high-fidelity load simulations as means for computationally inexpensive wind turbine load assessments; the so far best performing surrogate modelling approach in terms of balance between accuracy and computational cost has been the polynomial chaos expansion (PCE). Regarding the growing interest in advanced machine learning applications, the potential of using Artificial Neural-Network (ANN) based surrogate models for improved simplified load assessment is investigated in this study. Different ANN model architectures have been evaluated and compared to other types of surrogate models (PCE and quadratic response surface). The results show that a feedforward neural network with two hidden layers and 11 neurons per layer, trained with the Levenberg Marquardt backpropagation algorithm is able to estimate blade root flapwise damage-equivalent loads (DEL) more accurately and faster than a PCE trained on the same data set. Further research will focus on further model improvements by applying different training techniques, as well as expanding the work with more load components. IntroductionTypically wind turbines are designed for specific wind conditions which are specified in site classes by the IEC standards. When a turbine is placed at locations where a site-specific parameter exceeds these design conditions, site-specific load assessments including simulations over the whole design load base have to be carried out. As this procedure can become computationally expensive, several methods and procedures have been developed for simplifying load assessments based on statistical moments, multivariate regression models [1] and expansions using orthogonal polynomial basis [2]. Previous investigations comparing different surrogate models such as polynomial chaos expansion (PCE), universal kriging with polynomial chaos basis function and quadratic response surface, have shown that the PCE results in the best overall performance for the load estimation in terms of robustness, accuracy and computing time [3].Regarding the growing interest in advanced machine learning applications, the purpose of this study is to evaluate the potential of using models based on Artificial Neural Networks (ANNs) as a flexible and potentially better-performance alternative to the previously mentioned surrogate models that have been developed already. Therefore, different ANN models are trained for

No abstract